CN111858867A - Incomplete corpus completion method and device - Google Patents
Incomplete corpus completion method and device Download PDFInfo
- Publication number
- CN111858867A CN111858867A CN201910365369.5A CN201910365369A CN111858867A CN 111858867 A CN111858867 A CN 111858867A CN 201910365369 A CN201910365369 A CN 201910365369A CN 111858867 A CN111858867 A CN 111858867A
- Authority
- CN
- China
- Prior art keywords
- seed
- entity
- relation
- relational
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
Abstract
The invention discloses a completion method of incomplete linguistic data, which comprises the following steps: collecting user corpora; analyzing the corpus to obtain corresponding entity relation expression sentences; judging whether the entity relation expression sentence has figure entity incomplete or not; and when the entity relation expression sentence is judged to have the character entity defect, matching the entity relation expression sentence in a pre-stored seed relation library to complete the entity relation expression sentence. In addition, the invention also discloses a completion device of the incomplete corpus, which comprises: the corpus collection module is used for collecting user corpuses; the corpus analyzing module is used for analyzing the corpus to obtain corresponding entity relation expression sentences; a incomplete judgment module for judging whether the entity relation expression sentence has human entity incomplete; and the corpus complementing module is used for matching the entity relation expression sentences in a pre-stored seed relation library and complementing the entity relation expression sentences. By the method and the device, the incomplete linguistic data of the user can be supplemented, and the supplementing accuracy is high, so that the feedback can be correctly given to the user.
Description
Technical Field
The invention relates to the field of natural language processing, in particular to a completion method and a completion device for incomplete linguistic data.
Background
With the rapid development of network technology, intelligent terminals gradually design the aspects of people's life, and with the increasing maturity of artificial intelligence related technologies, the intelligent degree of various terminals is also higher and higher. The voice interaction is one of mainstream interaction modes in the intelligent terminal application, and great convenience is brought to the use of each user. For example, when the family education machine and the learning machine are used, children can ask various questions which are not understood by the family education machine and the learning machine through voice, and then the learning machine gives corresponding responses to the children.
For the question of the user, the intelligent device usually needs to collect the question information first and then completes the question corpus information so as to feed back the question information to the user in time and solve the question of the user, while the conventional completion mode mainly judges the completeness of the part of speech in the sentence after word segmentation, has low accuracy and is easy to cause the result of asking the answer.
Disclosure of Invention
In order to solve the above technical problems, the present invention provides a method and an apparatus for completing a incomplete corpus, and specifically, the technical scheme of the present invention is as follows:
on one hand, the invention discloses a completion method of incomplete linguistic data, which comprises the following steps:
collecting user corpora;
Analyzing the corpus to obtain corresponding entity relation expression sentences;
judging whether the entity relation expression sentence has figure entity incomplete or not;
and when the entity relation expression sentence is judged to have the character entity defect, matching the entity relation expression sentence in a pre-stored seed relation library to complete the entity relation expression sentence.
Preferably, matching the entity relationship expression sentence in a pre-stored seed relationship library, and completing the entity relationship expression sentence specifically includes:
searching a target seed in a seed relation library according to the existing figure entities in the entity relation expression sentence, wherein the target seed comprises a figure entity pair, and one figure entity in the figure entity pair is the existing figure entity in the entity relation expression sentence;
acquiring a corresponding target seed relation set according to the target seeds; the target seed relation set comprises a plurality of relation expressions;
matching the relational expressions contained in the entity relational expression sentence in the target seed relational set;
and when the relational expressions contained in the entity relational expression sentences are matched with the same relational expression in the target seed relational set, completing the entity relational expression sentences according to the character entity pairs contained in the seeds corresponding to the target seed relational set.
Preferably, before collecting the user corpus, the method further comprises:
constructing and storing a seed relational database, wherein the seed relational database comprises N seeds and corresponding seed relational sets; the method specifically comprises the following steps:
defining a seed containing a pair of human entities;
extracting the content containing the person entity pair in the seed from the collected document;
obtaining a seed relational expression according to the extracted content containing the seed entity pair;
establishing a seed relation set corresponding to the seeds according to the obtained seed relation expression;
and storing all the defined seeds and the corresponding seed relation sets in a seed relation library.
Preferably, the completion method of the incomplete corpus further includes:
when the entity relation expression sentence is judged to have no figure entity defect, extracting figure entity pairs and relation expressions contained in the entity relation sentence;
searching whether a target seed containing the extracted person entity pair exists in the seed relation library;
and when judging that the seed relational database does not have the target seed containing the extracted human entity pair, taking the extracted human entity pair as a new seed in the seed relational database, and establishing a seed relation set of the new seed, wherein the seed relation set of the new seed contains a relational expression contained in the entity relational expression sentence.
Preferably, the completion method of the incomplete corpus further includes:
when it is judged that the target seeds containing the extracted character entity pairs exist in the seed relation library, acquiring a target seed relation set corresponding to the target seeds;
matching the relational expressions contained in the entity relational expression sentence in the target seed relational set;
and when the relational expressions extracted from the entity relational expression sentences are not matched with the same relational expression in the target seed relational set, expanding the relational expressions contained in the entity relational expression sentences in the target seed relational set.
On the other hand, the invention also discloses a completion device of the incomplete corpus, which comprises the following components:
the corpus collection module is used for collecting user corpuses;
the corpus analyzing module is used for analyzing the corpus to obtain corresponding entity relation expression sentences;
the incomplete judgment module is used for judging whether the entity relation expression sentence has human entity incomplete;
and the corpus complementing module is used for matching the entity relation expression sentences in a pre-stored seed relation library when the entity relation expression sentences are judged to have character entity defects, and complementing the entity relation expression sentences.
Preferably, the corpus completion module includes:
a seed searching sub-module, configured to search a target seed in a seed relation library according to a person entity existing in the entity relation expression sentence, where the target seed includes a person entity pair, and one person entity in the person entity pair is the person entity existing in the entity relation expression sentence;
the relation matching submodule is used for acquiring a corresponding target seed relation set according to the target seeds; the target seed relation set comprises a plurality of relation expressions; matching the relational expressions contained in the entity relational expression sentence in the target seed relational set;
and the completion submodule is used for completing the entity relational expression sentence according to the character entity pair contained in the seed corresponding to the target seed relationship set when the relational expressions contained in the entity relational expression sentence are matched with the same relational expression in the target seed relationship set.
Preferably, the completion device for the incomplete corpus further includes:
the seed relation database comprises N seeds and corresponding seed relation sets; the building of the storage module specifically comprises:
A seed definition submodule for defining a seed containing a pair of human entities;
the content extraction submodule is used for extracting the content containing the character entity pairs in the seeds from the collected documents;
the expression generation submodule is used for obtaining a seed relational expression according to the extracted content containing the seed entity pair;
the set submodule is used for establishing a seed relationship set corresponding to the seeds according to the obtained seed relationship expression;
and the storage submodule is used for storing all the defined seeds and the seed relation sets corresponding to the seeds in the seed relation library.
Preferably, the completion device for the incomplete corpus further includes:
the extraction module is used for extracting the character entity pairs and the relational expressions from the entity relational sentences when the fact that the character entity defects do not exist in the entity relational expression sentences is judged; enabling the seed searching submodule to search whether a target seed containing the extracted person entity pair exists in the seed relation library;
and the expansion module is used for taking the extracted person entity pair as a new seed in the seed relational database and establishing a seed relationship set of the new seed when judging that the seed relational database does not have the target seed containing the extracted person entity pair, wherein the seed relationship set of the new seed contains a relational expression contained in the entity relational expression sentence.
Preferably, the completion device for the incomplete corpus further includes:
the seed searching sub-module is further configured to obtain a target seed relationship set corresponding to the target seed when it is determined that the target seed including the extracted person entity pair exists in the seed relationship library;
the relationship matching submodule is further used for matching the relational expressions contained in the entity relational expression sentence in the target seed relationship set;
and the expansion module is further used for expanding the relational expressions contained in the entity relational expression sentences in the target seed relationship set when the relational expressions extracted from the entity relational expression sentences are not matched with the same relational expressions in the target seed relationship set.
The invention has at least one technical effect as follows:
(1) after the user linguistic data are collected, the user linguistic data are analyzed, corresponding relational expression sentences are generated according to the user linguistic data, and then the pre-stored seed relational database is used for matching completion under the condition that the relational expression sentences are incomplete, so that the user can be responded according to the completed information. The method and the device can be used for complementing the incomplete linguistic data of the user, and the complementing accuracy is high, so that the feedback can be correctly given to the user.
(2) The method can not only complement the incomplete user corpus, but also add a new seed and a corresponding seed relation set in the seed relation library when judging that the user corpus is not incomplete and is not matched with the target seed, or add a relational expression to the target seed relation set when the target seed is matched but the relational expression is not matched; the seeds contained in the seed relation library and the content in the corresponding seed relation set are more and more comprehensive, and the incomplete linguistic data can be supplemented better in the follow-up process.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
FIG. 1 is a flowchart illustrating a completion method of a incomplete corpus according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating a completion method of the incomplete corpus according to another embodiment of the present invention;
FIG. 3 is a flowchart illustrating a completion method of the incomplete corpus according to another embodiment of the present invention;
FIG. 4 is a flowchart illustrating a completion method of the incomplete corpus according to another embodiment of the present invention;
FIG. 5 is a flowchart illustrating a completion method of the incomplete corpus according to another embodiment of the present invention;
FIG. 6 is a block diagram illustrating an embodiment of a completion apparatus for incomplete corpus according to the present invention;
FIG. 7 is a block diagram illustrating a structure of a completion apparatus for incomplete corpus according to another embodiment of the present invention;
FIG. 8 is a block diagram illustrating a completion apparatus for incomplete corpus according to another embodiment of the present invention.
Reference numerals:
10- -corpus collection module; 20- -corpus parsing module; 30- -deformity judgment module; 40- -corpus completion module; 41- -seed finding submodule; 42- -relationship matching submodule; 43- -completion submodule; 50- -building a storage module; 51- -seed definition submodule 51; 52- -content extraction submodule; 53- -expression generation submodule; 54- -Collection submodule; 55- -storage submodule; 60- -an extraction module; 70- -expansion module.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.
It will be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
For the sake of simplicity, the drawings only schematically show the parts relevant to the present invention, and they do not represent the actual structure as a product. In addition, in order to make the drawings concise and understandable, components having the same structure or function in some of the drawings are only schematically depicted, or only one of them is labeled. In this document, "one" means not only "only one" but also a case of "more than one".
It should be further understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
In particular implementations, the terminal devices described in embodiments of the present application include, but are not limited to, other portable devices such as mobile phones, laptop computers, family computers, or tablet computers having touch sensitive surfaces (e.g., touch screen displays and/or touch pads). It should also be understood that in some embodiments the terminal device is not a portable communication device, but is a desktop computer having a touch-sensitive surface (e.g., a touch screen display and/or touchpad).
In the discussion that follows, a terminal device that includes a display and a touch-sensitive surface is described. However, it should be understood that the terminal device may include one or more other physical user interface devices such as a physical keyboard, mouse, and/or joystick.
The terminal device supports various applications, such as one or more of the following: a drawing application, a presentation application, a network creation application, a word processing application, a disc burning application, a spreadsheet application, a gaming application, a telephone application, a video conferencing application, an email application, an instant messaging application, an exercise support application, a photo management application, a digital camera application, a digital video camera application, a Web browsing application, a digital music player application, and/or a digital video player application.
Various applications that may be executed on the terminal device may use at least one common physical user interface device, such as a touch-sensitive surface. One or more functions of the touch-sensitive surface and corresponding information displayed on the terminal can be adjusted and/or changed between applications and/or within respective applications. In this way, a common physical architecture (e.g., touch-sensitive surface) of the terminal can support various applications with user interfaces that are intuitive and transparent to the user.
In addition, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not intended to indicate or imply relative importance.
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following description will be made with reference to the accompanying drawings. It is obvious that the drawings in the following description are only some examples of the invention, and that for a person skilled in the art, other drawings and embodiments can be derived from them without inventive effort.
The invention discloses a completion method of incomplete linguistic data, which is implemented as shown in fig. 1 and can be applied to terminal equipment (such as a learning machine, a family education machine and the like, wherein the family education machine is used as a subject for convenient understanding in the embodiment, but the technical personnel in the field understand that the completion method of the incomplete linguistic data can also be applied to other terminal equipment as long as corresponding functions can be realized), and the completion method of the incomplete linguistic data comprises the following steps:
s101, collecting user corpora;
specifically, when a user learns, the learning machine is placed in front of the user, and the microphone is arranged on the learning machine, so that the voice of the user can be collected, and the user corpus can be obtained. For example, user corpora are collected: li wrote a poem to whom.
S102, analyzing the corpus to obtain corresponding entity relation expression sentences;
specifically, after the corpus of the user is obtained, semantic analysis can be performed on the corpus, and the included entity and entity relationship event are obtained according to the corpus information, so that a corresponding entity relationship expression sentence is generated: li write a poem to ____.
S103, judging whether the entity relation expression sentence has human entity incomplete or not;
specifically, it can be seen from the entity relationship expression sentence whether there is a person entity defect, for example, "li bai writes a poem and sends it to ____", it is obvious that there is a person entity object missing in the entity relationship expression sentence, that is, who the object to which the lie whites writes a poem is, cannot be known from the entity relationship expression sentence, and is defective, and it can be determined that there is a person entity defect in the entity relationship expression sentence.
And S104, when the entity relation expression sentence is judged to have the character entity defect, matching the entity relation expression sentence in a pre-stored seed relation library, and completing the entity relation expression sentence.
Specifically, when it is determined that there is a character entity defect in the entity relationship expression sentence, the defective entity needs to be complemented, and how to complement the character entity needs to be matched in a pre-stored seed relationship library, where the seed relationship library includes a large number of seeds, and each seed corresponds to a seed relationship set; each seed is in turn made up of a pair of entity pairs. When matching, specifically, a corresponding seed is searched in the seed relationship library, then a seed relationship set corresponding to the seed is searched, and if a matching relationship expression is searched in the seed relationship set, the missing entity relationship expression sentence can be completed according to the character entity pair in the seed of the seed relationship set, that is, the corpus of the user defect is completed. After completion, the family education machine can respond to the user according to the entity relationship expression sentence after completion.
In this embodiment, after the corpus of the user is collected, the corpus of the user is analyzed, a corresponding relational expression sentence is generated according to the analytic processing, and then the completion is matched by using the pre-stored seed relational database under the condition that the relational expression sentence is incomplete, so that the user can be responded according to the completed information.
Another embodiment of the method of the present invention, as shown in fig. 2, comprises:
s201, collecting user corpora;
s202, analyzing the corpus to obtain a corresponding entity relation expression sentence;
s203, judging whether the entity relation expression sentence has character entity incomplete or not;
s204, when it is judged that the entity relation expression sentence has the character entity defect, searching a target seed in a seed relation library according to the existing character entities in the entity relation expression sentence, wherein the target seed comprises a character entity pair, and one character entity in the character entity pair is the existing character entity in the entity relation expression sentence;
specifically, for example, the incomplete entity relationship expression sentence is that "a poem is written in Libai and sent to ____", and according to the existing character entity "Libai" in the expression, the seeds containing the character entity "Libai" can be found in the seed relationship library, for example, 3 target seeds are found: target seed 1 (the pair of human entities comprising: Libai, Dufu "); target seed 2 ("lilac, wanlon"); target seed 3 ("Libai, Menghanry").
S205, acquiring a corresponding target seed relation set according to the target seeds; the target seed relation set comprises a plurality of relation expressions;
specifically, after the target seed is obtained, a corresponding target seed relationship set is obtained according to the target seed. Because each seed has a unique corresponding seed relation set, the seed relation set contains a plurality of relational expressions which take the person entity pairs in the seed as objects. For example, after the three target seeds are found, the seed relationship set corresponding to each target seed can be found according to the target seeds, so as to obtain each relationship expression in each relationship set. Specifically, the target seeds and the relationship set schematic table thereof found in the seed relationship library are as follows:
TABLE 1 target seeds and relationship set mapping table
S206, matching the relational expressions contained in the entity relational expression sentence in the target seed relational set;
specifically, after obtaining the relational expressions in the target seed relationship set, the relational expressions included in the entity relational expression sentence are matched one by one. For example, the entity relationship expression sentence is: li is written with a poem to ____; it contains the relational expression: (Li Bai)Written a poem is sent to ___; firstly, matching the relation set of the target seed 1 with the relation set of the target seed 1, and matching the relation expression in the relation set of the target seed 1: x has written a poem to Y. After matching, matching can be continued for the following target seeds and the corresponding seed relationship sets, because multiple matching cases may occur. For example, the relational expression "X once written a poem and sent to Y" in the relational set of the target seed 2 is also successfully matched.
And S207, when the relational expressions contained in the entity relational expression sentence are matched with the same relational expression in the target seed relational set, completing the entity relational expression sentence according to the character entity pair contained in the seed corresponding to the target seed relational set.
Specifically, when the same relational expression is successfully matched in step S206, the corresponding seed is found according to the seed relationship set where the matched relational expression is located, and the incomplete entity relational expression sentence can be completed according to the character entity pair included in the seed. For example, relational expression sentences with the same semantics are matched in the target seed 1 relational set: "X once writes a poem to Y", and then according to the pair of person entities included in the target seed 1 relationship set: libai and Dufu, and further can fill in the previous incomplete entity relationship expression sentence: "Libai written a poem of Capehem to Dufu".
Similarly, because the relation expression sentences with the same semantics are also matched in the relation set of the target seed 2: "X once writes a poem to Y", and then according to the pair of person entities included in the target seed 2 relationship set: libai and Wanlun, and further can complement the incomplete entity relationship expression sentence before: ' Libai writes a poem to giveWanglon”。
According to the complemented entity relation expression sentence, a response can be given to the user, for example, the response is fed back to the user: "Libai written a capital poem to Dufu and Libai written a capital poem to Wanlun".
Another embodiment of the completion method of the incomplete corpus according to the present invention is based on any of the above embodiments, as shown in fig. 3, and includes:
s301, constructing and storing a seed relational database, wherein the seed relational database comprises N seeds and corresponding seed relationship sets;
s302, collecting user corpora;
s303, analyzing the corpus to obtain a corresponding entity relation expression sentence;
s304, judging whether the entity relation expression sentence has human entity incomplete or not;
s305, when the fact that the entity relation expression sentence has the character entity defect is judged, the entity relation expression sentence is matched in a pre-stored seed relation library, and the entity relation expression sentence is completed.
Wherein, the step S301 of constructing and storing the seed relationship library specifically includes:
s311, defining seeds containing the person entity pairs;
for example, the pair of human entities contained in seed 1 is defined as Libai, Dufu; defining the pair of human entities contained in the seed 2 as Libai and Wanlon; the pair of human entities contained in seed 3 is defined as: libai, bland.
S312, extracting the content of the character entity pair in the seed from the collected document;
specifically, after a large number of documents are collected, relevant content can be extracted from the documents according to the human entities contained in the seeds. For example, according to the person entity pair in seed 1, the following related contents are extracted:
(1) dou ever publicly shows that Du Dong Libai is good;
(2) libai and Dufu in Luocheng;
(3) libaizai write a poem to Dufu.
S313, obtaining a seed relational expression according to the extracted content containing the seed entity pair;
specifically, after the extracted content related to the seed entity pair is obtained, the related seed relational expression is obtained according to the extracted specific content. For example, the pair of human entities contained in seed 1 is Libai, Dufu; we denote lie by X and dupu by Y, and then obtain the corresponding seed expression according to the extracted related content:
(1) Dou ever publicly shows that Du Dong Libai is good;
the corresponding relational expression: x is publicly expressed, and is good at Y in Shandong
(2) Libai and Dufu in Luocheng;
the corresponding relational expression: x, Y Royal mountain
(3) Libaizai write a poem to Dufu.
The corresponding relational expression: x once written poem to Y
S314, establishing a seed relation set corresponding to the seeds according to the obtained seed relation expression;
then, the obtained seed relational expressions are collected, and a seed relational set corresponding to the seed is established. For example, a seed relationship set of seed 1 is established, and the seed relationship set includes three relational expressions: "X was published and indicated to be good at Y of Shandong"; "X, Y classed in Lowe"; and "X has written a poem to Y".
S315, storing all the defined seeds and the corresponding seed relation sets in a seed relation library.
Another embodiment of the method of the present invention is, on the basis of any of the above embodiments, to add an expansion step of the seed relationship library, and specifically, as shown in fig. 4, the method includes:
s401, collecting user corpora;
s402, analyzing the corpus to obtain a corresponding entity relation expression sentence;
S403, judging whether the entity relation expression sentence has human entity incomplete or not; if yes, go to step S404; otherwise, go to step S405;
s404, matching the entity relation expression sentence in a pre-stored seed relation library to complete the entity relation expression sentence;
s405, extracting the character entity pairs and the relational expressions from the entity relational sentences;
specifically, if the entity relation sentence is complete and there is no character entity incomplete, character entity pairs and relation expressions included in the entity relation sentence can be extracted from the entity relation sentence; for example, the entity relationship expression sentence obtained according to the user corpus is: 'Li Bai writes a poem to Menghan'. And judging that the entity relation expression sentence has no figure entity defects and belongs to a complete sentence, and then extracting figure entity pairs from the entity relation expression sentence: libai, Menghanry, and the expression of the entity relationship:(character entity a)Write a poem to give(character entity b)。
S406, searching whether a target seed containing the extracted person entity pair exists in the seed relation library; if not, go to step S407;
specifically, after the person entity pair and the corresponding entity relationship expression are extracted, whether corresponding seeds exist in the seed relationship library is searched according to the person entity pair, that is, whether seeds containing the pair of the person entities of 'Libai and peaceful' exist in the seed relationship library is searched.
S407, taking the extracted human entity pair as a new seed in the seed relation library, and establishing a seed relation set of the new seed, wherein the seed relation set of the new seed comprises a relation expression contained in the entity relation expression sentence.
Specifically, if the target seed including the extracted pair of human entities is not found in the seed relationship library, a new seed and a corresponding new seed relationship set need to be added in the seed relationship library. For example, if no seeds containing the pair of human entity "Libai, Mengtiany" are found in the existing seed relational database, a new seed needs to be added in the seed relational database, and the new seed is defined to contain the pair of human entity "Libai, Mengtiany"; and establishing a seed relation set of the new seed, wherein a relational expression contained in the seed relation set of the new seed is as follows: x writes a poem to Y. Wherein, the new seed relationship is centralized and X represents plum white and Y represents Mongolian.
Through the scheme of this embodiment, except can mending incomplete user's corpus, can also expand the seed relational database under the condition that judge that user's corpus is incomplete for the content in the seed that contains in the seed relational database and the seed relation set that corresponds is more and more comprehensive, also is convenient for follow-up better mends incomplete corpus.
Another embodiment of the method of the present invention is, on the basis of the above embodiment, adding an expansion process of a relational expression in a seed relationship set, specifically, as shown in fig. 5, including:
s501, collecting user corpora;
s502, analyzing the corpus to obtain a corresponding entity relation expression sentence;
s503, judging whether the entity relation expression sentence has human entity incomplete or not; if yes, go to step S504; otherwise, go to step S505;
s504, matching the entity relation expression sentence in a pre-stored seed relation library, and completing the entity relation expression sentence;
s505, extracting the character entity pairs and the relational expressions from the entity relational sentences;
s506, searching whether a target seed containing the extracted person entity pair exists in the seed relation library; if yes, go to step S508; if not, go to step S507;
and S507, taking the extracted character entity pair as a new seed in the seed relation library, and establishing a seed relation set of the new seed, wherein the seed relation set of the new seed comprises a relation expression contained in the entity relation expression sentence.
S508, acquiring a target seed relation set corresponding to the target seeds;
Specifically, when the target seed including the person entity pair extracted from the entity relationship expression relation sentence is found in the seed relationship library, the target seed relationship set corresponding to the target seed may be further obtained. For example, the entity relationship expression sentence "Libai writes a poem sent to Menghanry" includes the following pairs of character entities: plum white, menghanry; the contained entity relational expression is as follows:(character entity a)Write a poem to give(character entity b). If the seed with the 'Libai, Mengtiany' character entity pair is found in the seed relation library, the seed is the target seed, and then the corresponding target seed relation set can be obtained according to the target seed, so that the existing relation expression in the target seed relation set can be known.
S509, matching the relational expressions contained in the entity relational expression sentence in the target seed relational set;
specifically, an entity relational expression included in the entity relational expression sentence "(character entity a)Write a poem to give(character entity b)"match with each relational expression in the target seed relationship set. If two relational expressions are contained in the target seed relational set, specifically, X represents prunus and Y represents amazingly, the relational expressions contained in the target seed relational set are as follows:
(1) X and Y are mutually acquainted in the Yangyang;
(2) y is believed to invite X to the convergence of the involving yang.
After matching, the entity relational expression in the entity relational expression sentence is related to the relationship in the target seed relational set
The expression matching is unsuccessful.
S510, when the relational expressions extracted from the entity relational expression sentences are not matched to the same relational expression in the target seed relationship set, extending the relational expressions included in the entity relational expression sentences in the target seed relationship set.
Specifically, if the entity relational expression in the entity relational expression sentence is unsuccessfully matched with the relational expression in the target seed relationship set, the entity relational expression in the entity relational expression sentence may be expanded in the target seed relationship set, for example, the entity relational expression "li bai writes a poem to menghan" in the entity relational expression sentence, and the entity relational expression "X writes a poem to Y" in the entity relational expression sentence (likewise, X represents lie bai and Y represents menghan according to the target seed relationship set definition) is increased in the target seed relationship set. Then, the seed containing the pair of human entities "Libai, Menaman" now contains three relational expressions in the corresponding seed relation set:
(1) X and Y are mutually acquainted in the Yangyang;
(2) y once believes that X is involved in the convergence of the involving yang;
(3) x writes a poem to Y.
By the scheme of the embodiment, besides completion of the incomplete user corpus, the seeds in the seed relation library and the corresponding seed relation sets can be added when the fact that the user corpus is not incomplete and is not matched with the target seeds is judged, or the relational expressions are added to the target seed relation sets when the target seeds are matched but the relational expressions are not matched; the seeds contained in the seed relation library and the content in the corresponding seed relation set are more and more comprehensive, and the incomplete linguistic data can be supplemented better in the follow-up process.
Based on the same technical conception, the invention also discloses a completion device of the incomplete linguistic data, which can adopt the method of the invention to complete the incomplete linguistic data, thereby giving corresponding feedback to the user. Specifically, as shown in fig. 6, the completion apparatus for incomplete corpus of the present invention includes:
the corpus collection module 10 is used for collecting user corpora; specifically, the corpus collection module 10 may be implemented by a microphone or other voice collection device. For example, user corpora are collected: li white and who converged in lolo?
The corpus analyzing module 20 is configured to analyze the corpus to obtain corresponding entity relationship expression sentences; specifically, after the corpus of the user is obtained, semantic analysis can be performed on the corpus, and then the included entity and entity relationship event are obtained according to the corpus information, so that a corresponding entity relationship expression sentence is generated: libai and ____ phase Polylocheng.
A incomplete judgment module 30, configured to judge whether the entity relationship expression sentence has a human entity incomplete; specifically, the disability determination module 30 determines whether there is a human entity disability from the entity relationship expression sentence, for example, li bai and ____ jue locheng,
and the corpus complementing module 40 is configured to, when it is determined that the entity relationship expression sentence has the character entity defect, match the entity relationship expression sentence in a pre-stored seed relationship library, and complement the entity relationship expression sentence.
Specifically, when the incomplete judgment module 30 determines that there is a character entity incomplete in the entity relationship expression sentence, the incomplete entity needs to be complemented, and how to complement the incomplete entity needs to be matched in a pre-stored seed relationship library by the corpus complementing module 40, where the seed relationship library includes a plurality of seeds, and each seed corresponds to a seed relationship set; each seed is in turn made up of a pair of entity pairs. When matching, specifically, the corpus completion module 40 searches the corresponding seed in the seed relationship library, then searches the seed relationship set corresponding to the seed, and if the matched relationship expression is found in the seed relationship set, the missing entity relationship expression sentence can be completed according to the character entity pair in the seed of the seed relationship set.
In this embodiment, after the corpus collection module 10 collects the corpus of the user, the corpus analysis module 20 analyzes the corpus of the user, and generates a corresponding relational expression sentence according to the analytic processing, and then the incomplete judgment module 30 performs incomplete judgment on the relational expression sentence, and under the condition that it is judged that the relational expression sentence has character entity incomplete, the corpus completion module 40 matches completion by using a pre-stored seed relation library, so that the home education machine can respond to the user according to the completed information.
In another embodiment of the apparatus of the present invention, as shown in fig. 7, based on the above embodiment of the apparatus, the corpus completion module 40 includes:
a seed searching submodule 41, configured to search a target seed in a seed relation library according to a person entity existing in the entity relation expression sentence, where the target seed includes a person entity pair, and one person entity in the person entity pair is the person entity existing in the entity relation expression sentence;
specifically, for example, the incomplete expression sentence of the entity relationship is "lilai and ____ jugondolas", and the seed search sub-module 41 may search the seed containing the person entity "lilai" in the seed relationship library according to the existing person entity "lilai" in the expression.
The relation matching submodule 42 is configured to obtain a corresponding target seed relation set according to the target seed; the target seed relation set comprises a plurality of relation expressions; matching the relational expressions contained in the entity relational expression sentence in the target seed relational set; specifically, after the seed searching sub-module 41 obtains the target seed, the relationship matching sub-module 42 obtains the corresponding target seed relationship set according to the target seed. Because each seed has a unique corresponding seed relation set, the seed relation set contains a plurality of relational expressions which take the person entity pairs in the seed as objects.
A completion submodule 43, configured to match the relational expressions included in the entity relational expression sentence in the target seed relational setAnd when the same relational expression is matched, completing the entity relational expression sentence according to the character entity pair contained in the seeds corresponding to the target seed relation set. Specifically, after obtaining the relational expressions in the target seed relationship set, the relational expressions included in the entity relational expression sentence are matched one by one. For example, the entity relationship expression sentence is: libai and ____ are together in Luocheng, then it is further matched in the target seed relationship set to see if there is the same relational expression, which mainly means the semantic identity. If so, the matching is regarded as successful, and then the entity relational expression sentence can be complemented according to the human entity pair 'Libai, Dufu' contained in the target seed: "Libai and Dufu liquorCongregate in Luochen.
Preferably, on the basis of any one of the above embodiments of the apparatus, the apparatus for completing the incomplete corpus further includes:
the construction storage module 50 is used for constructing and storing a seed relation library, wherein the seed relation library comprises N seeds and corresponding seed relation sets; the building storage module 50 specifically includes:
a seed definition sub-module 51 for defining a seed containing a pair of human entities;
a content extraction sub-module 52, configured to extract content including the pair of human entities in the seed from the collected documents;
the expression generation submodule 53 is configured to obtain a seed relationship expression according to the extracted content including the seed entity pair;
the set submodule 54 is configured to establish a seed relationship set corresponding to the seed according to the obtained seed relationship expression;
and the storage submodule 55 is configured to store all the defined seeds and the respective corresponding seed relationship sets in the seed relationship library.
Specifically, the seed definition submodule 51 defines the pair of human entities contained in the seed 1 as lisu and dupu; defining the pair of human entities contained in the seed 2 as Libai and Wanlon; the pair of human entities contained in seed 3 is defined as: libai, bland. After a large number of documents are collected, the content extraction sub-module 52 may extract relevant content from the documents according to the human entities contained in the seeds. For example, according to the person entity pair in seed 1, the following related contents are extracted:
(1) Dou ever publicly shows that Du Dong Libai is good;
(2) libai and Dufu in Luocheng;
(3) libaizai write a poem to Dufu.
After the content extraction sub-module 52 obtains the extracted content related to the seed entity pair, the expression generation sub-module 53 obtains the related seed relationship expression according to the extracted specific content. For example, the pair of human entities contained in seed 1 is Libai, Dufu; we denote lie by X and dupu by Y, and then obtain the corresponding seed expression according to the extracted related content:
(3) dou ever publicly shows that Du Dong Libai is good;
the corresponding relational expression: x is publicly expressed, and is good at Y in Shandong
(4) Libai and Dufu in Luocheng;
the corresponding relational expression: x, Y Royal mountain
(3) Libaizai write a poem to Dufu.
The corresponding relational expression: x once written poem to Y
Then, the set submodule 54 sets the obtained seed relationship expressions together to establish a seed relationship set corresponding to the seed. For example, a seed relationship set of seed 1 is established, and the seed relationship set includes three relational expressions: "X was published and indicated to be good at Y of Shandong"; "X, Y classed in Lowe"; and "X has written a poem to Y".
Finally, the storage submodule 55 stores the previously defined seeds and the corresponding seed relationship sets in the seed relationship library.
In another embodiment of the apparatus according to the present invention, based on the above apparatus embodiment, as shown in fig. 8, the apparatus for completing the incomplete corpus further includes:
an extracting module 60, configured to extract a pair of people entities and a relationship expression included in the entity relationship sentence when it is determined that there is no people entity stub in the entity relationship expression sentence; enabling the seed searching submodule 41 to search whether a target seed containing the extracted person entity pair exists in the seed relation library;
and an expansion module 70, configured to, when it is determined that the seed relationship library does not have a target seed including the extracted human entity pair, take the extracted human entity pair as a new seed in the seed relationship library, and establish a seed relationship set of the new seed, where the seed relationship set of the new seed includes a relationship expression included in the entity relationship expression sentence.
Preferably, the completion device for the incomplete corpus further includes:
the seed searching sub-module 41 is further configured to, when it is determined that the target seed including the extracted person entity pair exists in the seed relationship library, obtain a target seed relationship set corresponding to the target seed;
The relationship matching sub-module 42 is further configured to match the relational expressions included in the entity relational expression sentence in the target seed relationship set;
the expansion module 70 is further configured to expand the relational expressions included in the entity relational expression sentence in the target seed relationship set when the relational expressions extracted from the entity relational expression sentence are not matched to the same relational expression in the target seed relationship set.
By the corpus complementing device of the embodiment, besides the incomplete user corpus can be complemented, when the fact that the user corpus is not incomplete is judged and the target seeds are not matched, the seeds in the seed relation library and the corresponding seed relation sets are added, or when the target seeds are matched but the relational expressions are not matched, the relational expressions are added to the target seed relation sets; the seeds contained in the seed relation library and the content in the corresponding seed relation set are more and more comprehensive, and the incomplete linguistic data can be supplemented better in the follow-up process.
The incomplete corpus completing method of the present invention corresponds to the incomplete corpus completing device of the present invention, and the technical details of the incomplete corpus completing method of the present invention are also applicable to the incomplete corpus completing device of the present invention, and are not repeated for reducing the repetition.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. A completion method for incomplete linguistic data is characterized by comprising the following steps:
Collecting user corpora;
analyzing the corpus to obtain corresponding entity relation expression sentences;
judging whether the entity relation expression sentence has figure entity incomplete or not;
and when the entity relation expression sentence is judged to have the character entity defect, matching the entity relation expression sentence in a pre-stored seed relation library to complete the entity relation expression sentence.
2. The method according to claim 1, wherein the matching of the entity-relationship expression sentences in a pre-stored seed relationship library comprises:
searching a target seed in a seed relation library according to the existing figure entities in the entity relation expression sentence, wherein the target seed comprises a figure entity pair, and one figure entity in the figure entity pair is the existing figure entity in the entity relation expression sentence;
acquiring a corresponding target seed relation set according to the target seeds; the target seed relation set comprises a plurality of relation expressions;
matching the relational expressions contained in the entity relational expression sentence in the target seed relational set;
And when the relational expressions contained in the entity relational expression sentences are matched with the same relational expression in the target seed relational set, completing the entity relational expression sentences according to the character entity pairs contained in the seeds corresponding to the target seed relational set.
3. The method according to claim 1, further comprising, before the collecting the user corpus:
constructing and storing a seed relational database, wherein the seed relational database comprises N seeds and corresponding seed relational sets; the method specifically comprises the following steps:
defining a seed containing a pair of human entities;
extracting the content containing the person entity pair in the seed from the collected document;
obtaining a seed relational expression according to the extracted content containing the seed entity pair;
establishing a seed relation set corresponding to the seeds according to the obtained seed relation expression;
and storing all the defined seeds and the corresponding seed relation sets in a seed relation library.
4. The method according to claim 1, further comprising:
when the entity relation expression sentence is judged to have no figure entity defect, extracting figure entity pairs and relation expressions contained in the entity relation sentence;
Searching whether a target seed containing the extracted person entity pair exists in the seed relation library;
and when judging that the seed relational database does not have the target seed containing the extracted human entity pair, taking the extracted human entity pair as a new seed in the seed relational database, and establishing a seed relation set of the new seed, wherein the seed relation set of the new seed contains a relational expression contained in the entity relational expression sentence.
5. The method according to claim 4, further comprising:
when it is judged that the target seeds containing the extracted character entity pairs exist in the seed relation library, acquiring a target seed relation set corresponding to the target seeds;
matching the relational expressions contained in the entity relational expression sentence in the target seed relational set;
and when the relational expressions extracted from the entity relational expression sentences are not matched with the same relational expression in the target seed relational set, expanding the relational expressions contained in the entity relational expression sentences in the target seed relational set.
6. The utility model provides a completion device of incomplete corpus which characterized in that includes:
The corpus collection module is used for collecting user corpuses;
the corpus analyzing module is used for analyzing the corpus to obtain corresponding entity relation expression sentences;
the incomplete judgment module is used for judging whether the entity relation expression sentence has human entity incomplete;
and the corpus complementing module is used for matching the entity relation expression sentences in a pre-stored seed relation library when the entity relation expression sentences are judged to have character entity defects, and complementing the entity relation expression sentences.
7. The apparatus according to claim 6, wherein the corpus completion module comprises:
a seed searching sub-module, configured to search a target seed in a seed relation library according to a person entity existing in the entity relation expression sentence, where the target seed includes a person entity pair, and one person entity in the person entity pair is the person entity existing in the entity relation expression sentence;
the relation matching submodule is used for acquiring a corresponding target seed relation set according to the target seeds; the target seed relation set comprises a plurality of relation expressions; matching the relational expressions contained in the entity relational expression sentence in the target seed relational set;
And the completion submodule is used for completing the entity relational expression sentence according to the character entity pair contained in the seed corresponding to the target seed relationship set when the relational expressions contained in the entity relational expression sentence are matched with the same relational expression in the target seed relationship set.
8. The apparatus for completing incomplete corpus as claimed in claim 6, further comprising:
the seed relation database comprises N seeds and corresponding seed relation sets; the building of the storage module specifically comprises:
a seed definition submodule for defining a seed containing a pair of human entities;
the content extraction submodule is used for extracting the content containing the character entity pairs in the seeds from the collected documents;
the expression generation submodule is used for obtaining a seed relational expression according to the extracted content containing the seed entity pair;
the set submodule is used for establishing a seed relationship set corresponding to the seeds according to the obtained seed relationship expression;
and the storage submodule is used for storing all the defined seeds and the seed relation sets corresponding to the seeds in the seed relation library.
9. The apparatus for completing incomplete corpus as claimed in claim 7, further comprising:
the extraction module is used for extracting the character entity pairs and the relational expressions from the entity relational sentences when the fact that the character entity defects do not exist in the entity relational expression sentences is judged; enabling the seed searching submodule to search whether a target seed containing the extracted person entity pair exists in the seed relation library;
and the expansion module is used for taking the extracted person entity pair as a new seed in the seed relational database and establishing a seed relationship set of the new seed when judging that the seed relational database does not have the target seed containing the extracted person entity pair, wherein the seed relationship set of the new seed contains a relational expression contained in the entity relational expression sentence.
10. The apparatus for completing incomplete corpus as claimed in claim 9, further comprising:
the seed searching sub-module is further configured to obtain a target seed relationship set corresponding to the target seed when it is determined that the target seed including the extracted person entity pair exists in the seed relationship library;
The relationship matching submodule is further used for matching the relational expressions contained in the entity relational expression sentence in the target seed relationship set;
the expansion module is further configured to expand the relational expressions included in the entity relational expression sentence in the target seed relationship set when the relational expressions extracted from the entity relational expression sentence are not matched with the same relational expression in the target seed relationship set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910365369.5A CN111858867A (en) | 2019-04-30 | 2019-04-30 | Incomplete corpus completion method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910365369.5A CN111858867A (en) | 2019-04-30 | 2019-04-30 | Incomplete corpus completion method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111858867A true CN111858867A (en) | 2020-10-30 |
Family
ID=72965850
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910365369.5A Pending CN111858867A (en) | 2019-04-30 | 2019-04-30 | Incomplete corpus completion method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111858867A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107784125A (en) * | 2017-11-24 | 2018-03-09 | 中国银行股份有限公司 | A kind of entity relation extraction method and device |
| CN108228877A (en) * | 2018-01-22 | 2018-06-29 | 北京师范大学 | Knowledge base complementing method and device based on study sort algorithm |
| CN108563637A (en) * | 2018-04-13 | 2018-09-21 | 北京理工大学 | A kind of sentence entity complementing method of fusion triple knowledge base |
| CN108874878A (en) * | 2018-05-03 | 2018-11-23 | 众安信息技术服务有限公司 | A kind of building system and method for knowledge mapping |
| CN109344231A (en) * | 2018-10-31 | 2019-02-15 | 广东小天才科技有限公司 | A kind of method and system of the semantic incomplete corpus of completion |
-
2019
- 2019-04-30 CN CN201910365369.5A patent/CN111858867A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107784125A (en) * | 2017-11-24 | 2018-03-09 | 中国银行股份有限公司 | A kind of entity relation extraction method and device |
| CN108228877A (en) * | 2018-01-22 | 2018-06-29 | 北京师范大学 | Knowledge base complementing method and device based on study sort algorithm |
| CN108563637A (en) * | 2018-04-13 | 2018-09-21 | 北京理工大学 | A kind of sentence entity complementing method of fusion triple knowledge base |
| CN108874878A (en) * | 2018-05-03 | 2018-11-23 | 众安信息技术服务有限公司 | A kind of building system and method for knowledge mapping |
| CN109344231A (en) * | 2018-10-31 | 2019-02-15 | 广东小天才科技有限公司 | A kind of method and system of the semantic incomplete corpus of completion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2019200437B2 (en) | A method to build an enterprise-specific knowledge graph | |
| CN109918676B (en) | Method and device for detecting intention regular expression and terminal equipment | |
| WO2020052405A1 (en) | Corpus annotation set generation method and apparatus, electronic device, and storage medium | |
| WO2021121198A1 (en) | Semantic similarity-based entity relation extraction method and apparatus, device and medium | |
| CN109582796A (en) | Generation method, device, equipment and the storage medium of enterprise's public sentiment event network | |
| WO2020151690A1 (en) | Statement generation method, device and equipment and storage medium | |
| CN110275962B (en) | Method and apparatus for outputting information | |
| EP3617910A1 (en) | Method and apparatus for displaying textual information | |
| CN111553138B (en) | Auxiliary writing method and device for standardizing content structure document | |
| CN112417846A (en) | Text automatic generation method and device, electronic equipment and storage medium | |
| WO2018149250A1 (en) | Chinese character skeleton code input method and system having suggestion screen interface | |
| CN113407536A (en) | Method and device for associating table data, terminal equipment and medium | |
| EP3992814A2 (en) | Method and apparatus for generating user interest profile, electronic device and storage medium | |
| WO2020232881A1 (en) | Text word segmentation method and apparatus | |
| CN117610649A (en) | Knowledge graph construction method and device, storage medium and electronic equipment | |
| CN110287270B (en) | Entity relationship mining method and equipment | |
| CN109977197B (en) | Electronic exercise processing method, device, equipment and storage medium | |
| WO2023078197A1 (en) | Classroom activity courseware producing method and apparatus, and storage medium and electronic device | |
| CN116629215A (en) | Teaching document generation method and device, electronic equipment and storage medium | |
| TWI530921B (en) | Cloud-based vocabulary learning system and method | |
| WO2023040516A1 (en) | Event integration method and apparatus, and electronic device, computer-readable storage medium and computer program product | |
| EP4109300A2 (en) | Method and apparatus for querying writing material, electronic device and storage medium | |
| CN111858867A (en) | Incomplete corpus completion method and device | |
| CN115757720A (en) | Project information searching method, device, equipment and medium based on knowledge graph | |
| CN110209895A (en) | Vector index method, apparatus and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |